414 MEASUREMENT OF ELECTROMOTIVE FORCES. 



but soon resumes its original electromotive force; if the substances 

 used are pure, the different specimens do not differ by more 

 than a thousandth. In this form there is no reason to fear any 

 diffusion, and the element once set up and sealed in the glass is 

 always ready for use. From the experiments of Lord Rayleigh,* 

 the electromotive force at 15 is i'435 volts. 



Rise of temperature diminishes the electromotive force of the 

 Latimer Clark element. The coefficients of variation obtained 

 by different physicists range from 0*00041 (Wright and Thomson), 

 to 0*00082 (Lord Rayleigh, Von Helmholtz) ; these differences 

 may arise from the mode of construction, but they may also 

 be due to an inexact determination of temperature, which is 

 difficult to ascertain from the form of the element itself, f By 

 introducing a thermometer in the interior of the instrument, Pellat 

 finds that the electromotive force of the Latimer Clark standard 

 is exactly represented between o and 25 by the formula 



E = E (i - 0-000781 /). 



1008. In order to fractionate the electromotive force of a 

 standard, the poles must be joined by a resistance which is 

 very great in comparison with its own ; from Ohm's law the 

 difference of potential of two points which comprise between them 

 any given fraction of this resistance is the same fraction of the 

 total electromotive force. 



Elliott's resistance boxes (927) enable us to make this 

 experiment in a very simple manner. The ordinary plugs being 

 placed at No. 9 of each dial, the standard is connected with the 

 extreme terminals, which introduces a resistance of 9999 ohms, 

 or of 10000 if a supplementary unit be added. If we place in 

 the plates of the dials two auxiliary plugs separated by a resistance 

 of n ohms, the difference of potential of these plugs is a fraction 

 equalling n ten thousandths of that of the terminals, or virtually of 

 the standard electromotive force. 



When the boxes are not such as to enable partial resistances 

 to be taken, two different boxes are joined end to end, the extreme 

 terminals of which are connected with the standard. If n units 

 of the first box are taken, and n' units of the second, the total 



* Lord RAYLEIGH. Phil. Trans. Roy. Soc. for 1884. Part n. p. 459. 

 t WRIGHT and THOMSON. Phil. Mag., [6], xvi. p. 33. 1883. HELMHOLTZ. 

 Berichte der Akad. zu Berlin, p. 26. 1882. 



